This paper provides statistics on quasi-monochromatic gravity wave events from 2017 to 2022, based on the observation of all-sky airglow imager. The propagation parameters of the waves were extracted using 2D Fourier transform, and the positions of wave sources were inferred with the ray tracing model.  In three case studies, the wave sources were related to jet streams, instead of deep convection. This paper is  based on abundant observation data and logically-organized, so I think this should be acceptable. However, I have two small questions as below:

1, In the abstract (line 1) and conclusion (line 463), the authors said there were 209 events found during the five years. But in line 114-115, the number of gravity wave cases changed to 64.  Why are there two different numbers?

2, How did the authors find out the total 209 (or 64)  gravity wave events from the airglow images captured in the five years? Have any methods been taken to avoid omissions and misjudgments?

General comments.

The paper considers about two hundred events of quasi-monochromatic atmospheric gravity waves (QMGWs) which were acquired over five years of observations in Southern Brazil, obtained with an OH all-sky imager. Gravity waves exhibited a seasonal dependence on the propagation direction with an anisotropic behavior. Typical characteristics of QMGWs are horizontal wavelengths of 10 - 55 km, observed periods of 5 – 74 minutes and observed phase speeds up to 100 m/s. A complimentary backward ray tracing analysis allowed the authors to reveal potential sources of QMGWs. I have found the paper to be interesting to the atmospheric community. At the same time, there are issues that are needed to explain and clarify. This is why a revision of the present paper is needed.

Specific comments.

Lines 2-3: “The observations were made using OH all-sky imagers hosted by the Southern Space Observatory (SSO)…”

How many all-sky imagers were used? One or several?

Lines 46-48: “Details on the observation mode (including the temporal resolution and integration time) and production of the final image can be seen elsewhere in Bageston et al. (2009) and Nyassor et al. (2021, 2022).”

Temporal and horizontal spatial resolutions are not “details” but the main parameters of the instrument, which should be presented here.

Line 66-67: “Afterward, the images are unwarped and mapped into the geographical coordinates. “

It is unclear how were images mapped into the geographic coordinates? In order to do it, one needs to make a total geometrical calibration (not only the alignment of the original airglow image to the geographical north) but solving the optical model of the camera for EACH pixel. A brief procedure of the geometrical calibration for each pixel is needed here.

Lines 76-77: “ Figure 1. Flowchart showing the procedures of airglow image processing and wave parameter estimation. The three stages describe image preprocessing and processing, spectral analysis, and wave parameters estimation procedure.”

In the flowchart and in section 3 “Methodology and Data Analysis”, I cannot see removals of the atmospheric background (not atmospheric extinction) and noise level of the sensor. Also, there is no information on the flat field correction of the sensor. How these issues have been treated in the image processing? These should be described.

Before presenting statistical results, it is worth presenting a figure demonstrating an example of an observed quasi-monochromatic gravity wave.

Something is wrong with the caption on the horizontal axis of Fig. 3c.

Line 153-154: “Similarly, most waves observed in the OH airglow images were visible and propagated for 2 - 3 hours.”

I do not understand this definition “duration of propagation of the waves in the OH images”. Does it mean that an observed wave package (a number of wave crests and troughs) did propagate for 2-3 hours in the field of view of the imager? Or is it something else? This should be explained.

Lines 406-407: “Also, Mastrantonio et al. (1976) showed that the gravity waves generated by tropospheric jet streams may have the ability to propagate vertically to the upper atmosphere, such as the ionosphere (Mastrantonio et al., 1976).”

Here it is worth citing two more papers:

Dalin et al. (2015) demonstrated a particular transient isolated gravity wave in the summer mesopause associated with the passage of an occluded front and/or the point of occlusion. The mechanism of the wave generation was likely due to strong horizontal wind shears at about 5 km altitude.

Dalin et al. (2016) demonstrated that gravity waves, observed in the summer mesopause, were associated with the upper tropospheric jet stream at altitudes 8–10 km.

Additional references:

Dalin, P., A. Pogoreltsev, N. Pertsev, V. Perminov, N. Shevchuk, A. Dubietis, M. Zalcik, S. Kulikov, A. Zadorozhny, D. Kudabayeva, A. Solodovnik, G. Salakhutdinov, I. Grigoryeva: Evidence of the formation of noctilucent clouds due to propagation of an isolated gravity wave caused by a tropospheric occluded front. Geophysical Research Letters, 42, 2037-2046, doi:10.1002/2014GL062776, 2015.

Dalin, P., N. Gavrilov, N. Pertsev, V. Perminov, A. Pogoreltsev, N. Shevchuk, A. Dubietis, P. Völger, M. Zalcik, A. Ling, S. Kulikov, A. Zadorozhny, G. Salakhutdinov, and I. Grigoryeva: A case study of long gravity wave crests in noctilucent clouds and their origin in the upper tropospheric jet stream. Journal of Geophysical Research - Atmospheres, 121, doi:10.1002/2016JD025422, 2016.